Lightnews — Scholar-powered news

Lukas Jung @lhdjung.bsky.social · 3d

Breaking: OSF finally got a proper download button.

@jamiecummins.bsky.social your prayers have been heard!

Screenshot of an OSF preprint page, now with a "Download preprint" button: https://osf.io/preprints/osf/zhm54_v1

1 1 14

Lukas Jung @lhdjung.bsky.social · 4d

So I read in a book that also portrayed a certain boisterous, loquacious statistics specialist.

1

Lukas Jung @lhdjung.bsky.social · 4d

I raise you this

1 3

Lukas Jung @lhdjung.bsky.social · 15d

Fair enough! 😀

1

Lukas Jung @lhdjung.bsky.social · 15d

You can run version$version.string to check, or packageVersion("base") if you only want the number.

1 1

Lukas Jung @lhdjung.bsky.social · 21d

Time had passed too, but I'm convinced language plays a role. Good writing is way easier in English than in German, at least.

1

Lukas Jung @lhdjung.bsky.social · 21d

I wrote my Bachelor's thesis in German, about a philosophical literature that was all in English. It was a pain, and the writing was not great. When I wrote my Master's in English, about a similar and partly overlapping English-only literature, it was much more enjoyable.

1 1

Lukas Jung @lhdjung.bsky.social · 23d

This can also be defined analogously to %in% itself:

`%not_in%` <- function(x, table) {
match(x, table, nomatch = 0L) == 0L
}

Note the == where %in% has >

2

Lukas Jung @lhdjung.bsky.social · Sep 8

Glad to see INSPECT-SR is out. A milestone in forensic metascience in medicine.

Jack Wilkinson @jdwilko.bsky.social · Sep 5

www.medrxiv.org/content/10.1...

INSPECT-SR: A tool for assessing trustworthiness of randomised controlled trials.

INSPECT-SR: a tool for assessing trustworthiness of randomised controlled trials

The integrity of evidence synthesis is threatened by problematic randomised controlled trials (RCTs). These are RCTs where there are serious concerns about the trustworthiness of the data or findings....

www.medrxiv.org

3

Lukas Jung @lhdjung.bsky.social · Aug 29

Another win for nominative determinism

1 2

Lukas Jung @lhdjung.bsky.social · Aug 19

"Replicators could not be expected to guess what needs to be done in order to test and demonstrate a phenomenon [...]. Our view is that in science everything should be clearly specified and should not involve unspecified procedures and assumptions, or require implicit knowledge."

Gilad Feldman @giladfeldman.bsky.social · Aug 19

Accepted for publication at Judgement and Decision Making:
"On process & value of direct close replications: Rejoinder to Shafir and Cheek (2024) commentary on Chandrashekar etal. (2021)"

Responding to original authors regarding our replication of Shafir (1993).

Details 👇🧵

1 8

Lukas Jung @lhdjung.bsky.social · Aug 16

I haven't had this issue, but you could try pak::pak() to install or update all packages required by your current project. The RStudio/Posit folks now recommend pak for package installation. pak.r-lib.org/reference/pa...

Install specified required packages — pak

Install the specified packages, or the ones required by the package or project in the current working directory.

pak.r-lib.org

1 2

Lukas Jung @lhdjung.bsky.social · Aug 15

The juxtaposition of science and "activism" is also a problem because the goal of all such activism is some vision of better science. By these standards, people are engaged in activism as opposed to science whenever they say, e.g., that more research is needed. Knowledge gain is always the point.

1 1